That seems flawed. Their sample was highly cherry picked, using people who are actually noted for being harassed in the first place. If you want to measure this accurately, you need a large RANDOM sample.
Yeah, it wouldn't be a surprise if this sample wasn't representative...but it isn't cherry-picked. To cherry-pick is to select more-or-less individual cases because they fit your hypothesis.
We don't know that the sample is representative...though we also don't know that it isn't.
Be interesting to see more studies like this, actually.
It's cherry picked in the sense that they went looking for people known for being harassed. That's pretty much guaranteed to give unrepresentative results.
3 Comments:
That seems flawed. Their sample was highly cherry picked, using people who are actually noted for being harassed in the first place. If you want to measure this accurately, you need a large RANDOM sample.
Yeah, it wouldn't be a surprise if this sample wasn't representative...but it isn't cherry-picked. To cherry-pick is to select more-or-less individual cases because they fit your hypothesis.
We don't know that the sample is representative...though we also don't know that it isn't.
Be interesting to see more studies like this, actually.
It's cherry picked in the sense that they went looking for people known for being harassed. That's pretty much guaranteed to give unrepresentative results.
Post a Comment
Subscribe to Post Comments [Atom]
<< Home